Re: Improve join selectivity estimation using extended statistics

> On 19 Jul 2021, at 12:52, Ibrar Ahmed wrote: > The patch does not compile, and needs your attention. > > https://cirrus-ci.com/task/6397726985289728 > > > clausesel.c:74:28: error: too few arguments to function > ‘choose_best_statistics’ >

Re: Improve join selectivity estimation using extended statistics

On Mon, Mar 15, 2021 at 8:42 PM Konstantin Knizhnik < k.knizh...@postgrespro.ru> wrote: > > > On 11.03.2021 03:47, Tomas Vondra wrote: > > Hi Konstantin, > > > > Thanks for working on this! Using extended statistics to improve join > > cardinality estimates was definitely on my radar, and this

Re: Improve join selectivity estimation using extended statistics

On 11.03.2021 03:47, Tomas Vondra wrote: Hi Konstantin, Thanks for working on this! Using extended statistics to improve join cardinality estimates was definitely on my radar, and this patch seems like a good start. I had two basic ideas about how we might improve join estimates: (a) use

Re: Improve join selectivity estimation using extended statistics

Hi Konstantin, Thanks for working on this! Using extended statistics to improve join cardinality estimates was definitely on my radar, and this patch seems like a good start. I had two basic ideas about how we might improve join estimates: (a) use per-table extended statistics to estimate join

Improve join selectivity estimation using extended statistics

Original thread is: https://www.postgresql.org/message-id/flat/196f1e1a-5464-ed07-ab3c-0c9920564af7%40postgrespro.ru Following Yugo's advice, I have splitted this patch into two: 1. Extending auto_explain extension to generate extended statistics in case of bad selectivity estimation. 2. Taken